【Running posture training shoes】No.007-Self-generating function research
[Copy link]
This post was last edited by sipower on 2021-6-27 03:09
The previous article introduced the design process of the Bluetooth communication program , and this one introduces the research on the self-power generation function. This may also be the part that everyone is most interested in.
The biggest feature of the RSL10 chip is that it saves power. The official gave an example application of using the ZF self-generating module to make a light switch. When I applied for the competition, I thought of using this module to make something, and finally decided to make running posture training shoes . Unfortunately, I didn't apply for the ZF self-generating module sample in the end , so I had to find a substitute on my own.
When looking for strange things, you can't do without Taobao APP. I found the following two self-generating things on it.
Figure 1: Self-generating things
The one on the left looks more professional, but it is more expensive. The module plus shipping costs more than 70, and I don’t know if it can be used. The one on the right doesn’t seem to be suitable for use on the feet, and it is also very large. Then the clever Taobao pushed a lot of self-powered products, among which the most popular ones are doorbells. Then I searched and searched, and finally I found the doorbell below. It is not only cheap, but even if it can’t be used in the end, it can still make use of its residual heat by installing it on the door. This is really good!
Figure 2, the selected self-generating doorbell
Place the order, pay, and it's done! What follows is a long wait --- NO, NO, NO. After a short wait, the product arrived a day later. Poor old father like me, I couldn't play with it for a few times before the kids snatched it away to have fun.
Figure 3, real photo
If you want to use the self-generating module inside, you naturally have to disassemble it first.
Figure 4, disassembly photo
The general principle of the self-generating doorbell is: the electric energy generated by pressing the power generation module is passed through the rectifier and filter circuit of the sending module to obtain stable electric energy. During this period, the chip of the wireless module is started and the coded radio signal is sent through the antenna. After the radio receiving module receives the coded radio signal, it decodes it and sends the ringtone if the code is consistent. The power generation principle is shown in the figure below.
Figure 5, Power generation principle
If you are interested and want to know more, you can refer to the following link. The author's introduction is very detailed and clear.
http://www.51hei.com/bbs/dpj-180596-1.html
Through the research on the wiring of this circuit board, I formulated my modification plan. I need to cut the VCC wiring, remove the power consumption of the original circuit, and lead out the power supply for my circuit. As shown in the figure below.
Figure 6, Transformation Circuit
During the wiring change, I somehow ended up burning the transmitter chip at the back, and the power and ground were short-circuited. I was thinking about letting it work, but it seems hopeless. I have to be careful later, so as not to burn my precious development board.
Since this board already comes with a rectifier bridge, voltage regulator, freewheeling capacitor and other components, I can just use them directly. Fortunately, I bought the finished product. If I bought the module separately, I would have to figure out this part of the circuit myself. I used a 10KΩ resistor as a load, and the waveform generated by continuous pressing is as shown below.
Figure 7, Test power
As you can see in the picture, the voltage can basically be stabilized above 4V, and it can generate about 0.4mA current. If the LDO is stepped down to 3.3V, I can get about 1mW of power. According to sending a packet about once a second, it should be enough. The next step is to power on the board. The board with the wires connected is as shown below.
Figure 8: Board with wires connected
Before powering on, be sure to select the power jumper correctly, especially to completely power off the J-link debug interface chip, otherwise all the energy will be absorbed by it. See the figure below.
Figure 9, Jumper Selection
The first test was not successful. As can be seen in the figure below, the energy was quickly attenuated.
Figure 10, Load waveform
After carefully observing the board, I found that the user LED was flashing. I had set it to be off in the program, but why was it still flashing? Could it be caused by the uncertainty of the IO port during the program initialization? ! I will first solder off its current limiting resistor, so that no matter what happens to the IO port, there will be no current.
Figure 11, soldering off the current limiting resistor
After another test, the waveform displayed by the oscilloscope is almost the same as the previous 10K load resistor. Under my continuous crazy output, the Bluetooth finally works normally and the mobile phone APP can also receive data. As shown in the figure below.
Figure 12, Connection successful
As you can see from the video below, the circuit can only work properly if I keep running crazy. In practice, this state can only be achieved in a 100-meter sprint.
Video 1, my crazy output
The reason is that the board consumes too much power, which causes power shortage. There are mainly the following reasons.
1. There is a part of the board that is not powered off. As shown in the figure below, these chips are not actually used, but the development board only has this one. If it is soldered off, it may damage the board and it will be useless later. The LDO also only uses the NCP551 that comes with the development board, instead of the ultra-low Iq LDO such as NCP170.
Figure 13: ICs that consume extra power
Second, the program is not carefully optimized. Due to my limited understanding of this chip, I only found the simplest routines to modify the code, and did not perform detailed power consumption adjustment. Many low-power functions were not used.
3. The self-generating module is also relatively simple and has low power. The price of the entire doorbell product is not as high as that of a more professional module, not to mention compared with the products of international manufacturers such as ZF.
In summary, this work is basically completed by the end of this article. Although it is relatively simple, the basic functions are realized, and these experiments also prove the power of the RSL10 chip. If you really want to make a good product, you need to make very careful choices, precise adjustments, and rigorous verifications in device selection, circuit optimization, and program design to achieve an excellent user experience. Of course, the money and time spent are not something that an amateur player like me can afford. I hope that my work can serve as a starting point for more users to be inspired to make functional and easy-to-use products.
Later I will continue to study the design of the upper computer APP, but this is an area I am not particularly good at. If I can come up with something presentable, I will continue to post. If I can't, I will not embarrass myself anymore.
Finally, I would like to thank all the moderators and experts in the forum for their help, and thank ON Semiconductor and Avnet for giving me this opportunity to showcase my work. I hope everyone can learn something from this event and receive a satisfactory award.
Program source code:
peripheral_server.zip
(659.89 KB, downloads: 6)
| 
提升卡
变色卡
千斤顶
京公网安备 11010802033920号
